Can your voice reveal hidden health conditions? According to a growing body of research, it just might. A recent Nature article by Neil Savage explores how scientists are using artificial intelligence (AI) to analyse voice recordings in search of early biomarkers for illnesses such as heart conditions, diabetes, Parkinson’s disease, Alzheimer’s and mental illnesses.
Voice is a complex, sensitive instrument, shaped not only by our lungs and vocal cords, but also by our neurological, psychiatric and emotional states. Subtle changes in pitch, rhythm, tone or articulation can signal underlying conditions long before traditional clinical symptoms emerge.
Laryngologist Yael Bensoussan, for instance, detected a serious heart condition simply by hearing how breathless a patient sounded. Now, she and others are developing AI models capable of picking up similarly telling signs, even those imperceptible to the human ear. These models could help non-specialists detect early health issues, offer non-invasive screening options via mobile phones, and enhance remote patient monitoring.
From menopause to mental health, diabetes to pulmonary disease, researchers are uncovering how physiological changes influence the sound of our speech. The TRUSTING project and other similar initiatives are at the forefront of this movement, using speech data to identify early relapse in psychosis and to promote more personalised, proactive care.
TRUSTING’s Brita Elvevåg, cognitive neuroscientist at UiT – the Arctic University of Norway in Tromsø, also was also featured in the Nature article, alongside postdoctoral researchers Musarrat Hussain and Enrico Tedeschi. Their work highlights the value of combining clinical insight and computational tools to identify early warning signs in serious mental illness.
As AI evolves, voice biomarkers may soon play a central role in diagnostics and preventive care. But challenges remain, including the need for large, diverse datasets and ethical safeguards. Still, the potential is clear: listening carefully, with the help of technology, may save lives.
Read the full article here: doi: https://doi.org/10.1038/d41586-025-01598-8
Stay Connected
For more updates, insights, and news about TRUSTING’s progress, visit TRUSTING Project or follow us on social media:
- X: https://x.com/Trusting_EU
- LinkedIn: https://www.linkedin.com/company/trusting-eu
- Sign up for the yearly newsletter: shorturl.at/SvxFN
